Importance of 3D convolution and physics on a deep learning coastal fog model
Importance of 3D convolution and physics on a deep learning coastal fog model
复制标题
DOI：
10.1016/j.envsoft.2022.105424
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
Hamid Kamangir;Evan Krell;Waylon Collins;Scott A. King;P. Tissot
中科院分区：
文献类型：
--
作者：
Hamid Kamangir;Evan Krell;Waylon Collins;Scott A. King;P. Tissot
危险大气现象的预报通常具有挑战性。人工智能（AI）模型已应用于大气科学问题。模型复杂性提供了量化模型架构组件重要性的动机。我们研究了为大大气数据设计的 FogNet 模型组件的相对重要性：1）3D 与 2D 卷积，2）基于物理的气象输入特征分组和排序，3）不同的基于 CNN 的辅助特征学习模块和 4）并行与顺序空间变量特征学习。我们通过预测沿海雾这一复杂的时空动态过程来研究这些 CNN 架构特征的相对重要性。我们使用四种可解释的人工智能技术来更好地理解输入特征的贡献。实验结果表明，基于 3D-CNN 的模型比 2D-CNN 更好地捕捉雾预测过程的复杂性。我们还表明，基于物理的特征分组以及它们输入 CNN 的顺序会显着影响性能。
The forecasting of hazardous atmospheric phenomena is often challenging. Artificial intelligence (AI) models have been applied to atmospheric science problems. Model complexity provides a motivation to quantify the importance of model architecture components. We studied the relative importance of the components of the FogNet model that was designed for big atmospheric data: 1) 3D versus 2D convolution, 2) physics-based grouping and ordering of meteorological input features, 3) different auxiliary CNN-based feature learning modules and 4) parallel versus sequential spatial-variable-wise feature learning. We investigate the relative importance of these CNN architectural features by predicting coastal fog,a complex spatiotemporal dynamical process. We use four explainable AI techniques to better understand input feature contributions. The results of the experiments demonstrate that 3D-CNN based models better capture the complexity of the fog prediction process than the 2D-CNNs. We also show that physics-based feature grouping, and the order in which they are fed into the CNNs, significantly impacts performance.